Community-acquired pneumonia (CAP) is a frequently observed, costly health issue causing significant morbidity and mortality. The incidence of CAP requiring hospitalization is about 25–30 cases per 10,000 adults and, in the USA, constitutes the seventh most frequent cause of all-cause death. Among infectious diseases, CAP is the most frequent cause of hospitalization and mortality in industrialized countries.1–3

In hospitalized patients, the mortality rate from CAP is around 10%.4–7 However, this rate varies depending on the hospital unit and the prognosis is worse for patients requiring treatment in an intensive care unit (ICU).

A large number of studies focus on the causal link between mortality and CAP, most of them analyzing in-hospital and short-term mortality.8,9 Fewer studies, however, have focused on the association between CAP and long-term mortality. At follow-up, pneumonia patients have displayed lower rates of survival and more frequent all-cause hospitalization, emergency department admissions and CAP-related visits compared to age and gender-matched control subjects without pneumonia.10 One-year mortality rate for patients with CAP is 17–40%, with increasing rates in the longer-term, independent of demographics and comorbid conditions.11–13 Recent work has sought to understand the reasons underlying the shorter survival of these patients and the incidence, factors, and predictors of long-term outcome. Long-term prognostic factors to be considered in CAP include advanced age, male gender, black race, health-care associated pneumonia, and chronic comorbid illnesses.5 To date, only a few systematic data evaluating long-term outcomes for these patients have been reported.

The objective of the present study is to investigate the short and long-term outcome in hospitalized patients with CAP and to identify the predictive factors of mortality.

This study was designed as a retrospective, multicenter, observational cohort study and was conducted in eight hospitals (seven university hospitals and one training and research hospital) within Turkey. The study complied with the ethical principles of the Declaration of Helsinki and was approved by the local Ethics Committee.

Of all the 785 patients included in the study, 68% were male and the mean age was 67±16 (18–92). A cough, sputum, and fever were the most common presenting symptoms. Overall, 82% of the patients had been diagnosed with at least one additional disease including: chronic obstructive pulmonary disease (COPD), asthma, congestive heart failure (CHF), coronary heart disease (CHD), diabetes mellitus (DM), malignancy, chronic renal disease (CRD), chronic liver disease and cerebrovascular disease. The average PSI and CURB-65 scores were 99.1±35.9 (15–243) and 2.16±0.98 (0–5), respectively.

Radiological findings included alveolar consolidation in 73% of the patients and pleural effusion in 21%. Multilobar infiltration was noted in 29% of patients (Table 1).

The median duration of follow-up was 61.2±11.8 (37–90) months. During this period, 428 (54.5%) patients died and the median survival of all patients was 50.1±4.2 months.

With respect to all-cause mortality, 30-day and 90-day mortality were 9.2% and 16.8%, respectively. Cumulative percentage mortality from one to five years post admission was 29.9%, 37.7%, 42.8%, 57.7%, and 74.8%, respectively (Table 2, Fig. 1).

Figure 1.

Kaplan–Meier survival curve for all patients.

(0.05MB).

The present study confirms the increased mortality rates following pneumonia. In terms of short-term mortality, aside from PSI and CURB-65 scores, higher ratios of serum BUN/albumin and radiological multilobar involvement were found to be significant predictors of a worse prognosis. Independent predictors for long-term mortality were advanced age, Charlson comorbidity score, absence of fever, higher BUN/albumin ratios and lower ALT levels. BUN/albumin ratios and ALT levels in this large study group are described for the first time as primary predictors of long-term outcome.

The 30-day mortality of hospitalized patients with CAP was verified by this study to be approximately 8.7–12%.7,8,18 Long-term survival rates vary widely in the literature. One-year mortality is reported to be between 17% and 40.9%.11,13 Two-year, 3-year and 5-year mortalities are reported as 27%, 31%, and 43%, respectively.13,18,19 Johnstone et al. reported 3.8-year mortality as 53%.7 Such variability could be related to different study groups, and/or exclusion of short-term mortality patients from the long-term mortality rate analysis. The current study calculated mortality rates for all the patients and has found survival rates to be low, in common with the existing literature.

The absence of fever at presentation was found to have a significant impact on mortality. The presence of fever had also been found to be protective in other studies.20 PORT study concluded that the lack of feeling feverish was related to poorer outcome compared to age-matched controls.18 We believe that infections causing fever reflect a better immune response and for effective host defense, hence leading to a better outcome. Higher respiratory rates may be related to more severe pneumonia. The relationship between long-term survival and respiratory rate would benefit from further research.

Previous studies evaluating causes of death have shown that the absolute CAP-related mortality rate is low whereas malignancy and diseases of the circulatory and respiratory systems are the most common causes of death.10,13 At follow-up, COPD, malignancy, neurodegenerative disorders and cardiovascular diseases were shown to increase mortality.5,13,20–22 Charlson index score has been correlated independently with the long-term outcome.18,23 Similarly, in the current study, Charlson score was found to be an independent predictor of long-term mortality. Furthermore, the score was also found to be a predictor of short-term outcome.

Lower albumin and higher levels of serum urea have been shown to increase mortality.24,25 Elevated serum BUN/albumin ratio was firstly described as a predictor of 28-day mortality by Ugajin et al. They also concluded that BUN/albumin ratio predicted the requirement for intensive care treatment in CAP.14 Contrary to this study, Akpinar et al. found that this ratio was able to predict the ICU admission but not 30-day mortality.26 However, the sample size was small in both of these studies. In the current study, a higher BUN/albumin ratio predicted a worse 30-day outcome. Additionally, this ratio predicted long-term survival in an independent manner. To the best of our knowledge, BUN/albumin ratio is described as a predictor of long-term prognosis for the first time.

Previous studies have shown that abnormal liver function tests, particularly raised ALT, increase in-hospital mortality.7,27 However, in the present study, no such relationship was discovered between liver function tests and short-term survival. Conversely, higher levels of ALT were protective for long-term survival. To the best of our knowledge, liver function tests have not been evaluated in terms of long-term outcome. Further studies are needed to cast light on this relationship.

In accordance with the literature, PSI and CURB-65 scores were found to be predictors of short-term mortality.14,28 PSI and CURB-65 scores have also been reported to be predictors of long-term mortality.18,28

The following limitations to this paper should be noted. First, it was a retrospective study. Second, an analysis of cause of death was not performed. Third, outpatients and ICU admissions were not included in the study. The strength of this study, on the other hand, is the large study sample with a well-documented database. Additionally, it is a multicenter study reflecting results which geographically covered the whole country.

In conclusion, long-term mortality following hospitalization for CAP is high. Patients with advanced age, higher Charlson score, higher respiratory rates and lower fever have a significantly increased chance of late mortality. Baseline BUN/albumin ratios and ALT levels are significant predictors of survival. Further interventions and closer monitoring would be necessary for such patient subgroups.

There is no conflict of interest related to this paper. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Informed consent was obtained from the patients included in the TURCAP database.